Cabinet Office issues ethics framework for data science

Written by Web News Editor on . Posted in News

The government has issued draft guidance (PDF) for policy makers and data scientists utilising government data, to help them navigate the legal and ethical issues that they might come up against. 

The ‘1.0’ version of the guidance brings together relevant laws, such as data protection and intellectual property laws, and gives six basic principles of best practice in order to give those working with government data 'the confidence to innovate’.

The six principles are:

  1. Start with clear user need and public benefit
  2. Use data and tools which have the minimum intrusion necessary
  3. Create robust data science models
  4. Be alert to public perceptions
  5. Be as open and accountable as possible
  6. Keep data secure.

The guidance adds that ‘the public benefit of doing the project needs to be balanced against the risk of doing so’ and then goes on to explain each principle in more detail, giving case studies as examples of each. The guidance also includes a Privacy Impact Assessment to help those working on data science projects identify what the risks to privacy might be and how this risk could be minimised.

Market research company Ipsos MORI was commissioned to assess public opinion on how government should use data science, and produced a report, Data Science Ethics Dialogue. Its research so far has shown that once the aims of a specific project are understood, public concerns about risk is usually measured against the aims of a project and its potential benefits.

The Royal Statistical Society views the new Data Science Ethics Framework as a step forward in utilising data for public good. 'The RSS has published research showing there is a "data trust deficit" and so we welcome this new framework as one mechanism which can maintain trustworthiness,’ says RSS executive director Hetan Shah. 'The deliberative research conducted by Ipsos MORI shows that public views on a data science project depends very much on context and impact, and so a case by case approach is important.' 

Further work on the framework will come from more research with the public on attitudes to data science ethics, to be conducted by Ipsos MORI, as well as feedback from other sources such as experts, civil servants and other interested parties. Details of how to get in touch are given at the bottom of this blog post.

Big Data Open Data Data privacy Data sharing